A serendipitous finding led us to develop a novel "modulatory" combination therapy for HIV infection. Dipyridamole (DPM; Persantin), a widely used, inexpensive cardiovascular agent, is also a potent inhibitor of nucleoside transport. We found that DPM potentiates the activity of AzT against HIV-1 in cultured human monocyte/macrophages and stimulated T-cells. In human T-lymphoblastoid cells (CEM-SS), DPM potentiates the anti-HIV activity and simultaneously protects the cells against AZT. DPM does not potentiate AZT's cytotoxicity for human bone marrow progenitors in the CFU-GM assay. Taken together, these findings suggested that DPM might increase the therapeutic index of AZT in vivo and/or decrease the cost of antiviral therapy. We are collaborating with others on phase I and phase Il clinical trials of the AZT/DPM combination. Aspects under study by our group include: 1. Mechanism of action: DPM blocks cellular uptake of physiological nucleosides but not of AZT and ddC. The potentiation of AZT (and ddC) may thus result, in part, from decreased influx of the nucleosides that compete with AZT for viral reverse transcriptase. 2. Molecular structure: Based on crystallographic findings, we computed conformer structures for DPM. Quantitative structure-activity relationships (3D-QSAR) analyses were then done to predict which features of nucleoside transport-inhibiting molecules are required for activity. 3. Molecular biology: We are trying to clone the equilibrative nucleoside transporter, a molecular target for DPM, but so far without success. 4. Clinical trials: A small clinical trial of AZT/DPM at the Henry Jackson Foundation (in collaboration with our group and the Dept of Clinical Pharm., Johns Hopkins U.) met phase I goals of maximum tolerated dose determination and pharmacokinetic characterization. A phase II protocol is under review. 5. General analysis of combination therapy: No published algorithm or computer package was adequate for analysis of data on antiviral drug combinations (including our own experiments). Hence,we have developed a new approach. The new concepts and computer program package (COMBO) are being used in collaboration with other groups at NIH and elsewhere to analyze data and design experiments on therapy of cancer and AIDS. Supported in part by a grant from the NIH Intramural AIDS Targeted Antivirals Program.